Assessment of thepharmacology andtolerability of PF-04457845,an irreversible inhibitor offatty acid amide hydrolase-1,in healthy subjectsGai Ling Li,1 Helen Winter,2 Rosalin Arends,2 Gary W. Jay,2 Vu Le,3

Tim Young1 & John P. Huggins1

1Pfizer Worldwide Research and Development, Sandwich, UK, 2Pfizer Worldwide Research and

Development, New London, CT and 3Pfizer Worldwide Research and Development, St Louis, MO, USA

CorrespondenceDr John P. Huggins, Pfizer WorldwideResearch and Development, SandwichCT13 9NJ, UK.Tel.: 01304 379010E-mail: johnphuggins@btinternet.com----------------------------------------------------------------------

Keywordsfatty acid amide hydrolase-1 (FAAH1),fatty acid amide, irreversible inhibitor,PF-04457845, pharmacokinetics----------------------------------------------------------------------

Received14 July 2011

Accepted26 October 2011

Accepted ArticlePublished Online1 November 2011

WHAT IS ALREADY KNOWN ABOUTTHIS SUBJECT• Inhibition of fatty acid amide hydrolase-1

(FAAH1) and the subsequent elevation offatty acid amides has been proposed as astrategy to induce the analgesic propertiesof cannabinoids without the accompanyingnegative side effects such as impairment incognition, motor control and predispositionto psychoses. PF-04457845 is a potent andselective irreversible FAAH1 inhibitor whichhas been shown to elevate fatty acid amideconcentrations in animal models and induceresponses in behavioural models suggestiveof analgesia.

WHAT THIS STUDY ADDS• This study is the first to investigate a FAAH1

inhibitor in humans. PF-04457845 is welltolerated following single and multipledosing to healthy volunteers and haspharmacokinetic and pharmacodynamicproperties that make nearly completeinhibition of FAAH1 possible with once dailydosing.

AIMSTo evaluate the pharmacology and tolerability of PF-04457845, anorally available fatty acid amide hydrolase-1 (FAAH1) inhibitor, inhealthy subjects.

METHODSDouble-blind, randomized, placebo-controlled single and multiplerising dose studies and an open-label, randomized, food effect studywere conducted. Plasma and urine PF-04457845 concentrations, plasmafatty acid amide concentrations and FAAH1 activity in humanleucocytes were measured. Tolerability, including effects on cognitivefunction, were assessed.

RESULTSPF-04457845 was rapidly absorbed (median tmax 0.5–1.2 h). Exposureincreased supraproportionally to dose from 0.1 to 10 mg andproportionally between 10 and 40 mg single doses. Thepharmacokinetics appeared dose proportional following 14 days oncedaily dosing between 0.5 and 8 mg. Steady-state was achieved by day7. Less than 0.1% of the dose was excreted in urine. Food had no effecton PF-04457845 pharmacokinetics. FAAH1 activity was almostcompletely inhibited (>97%) following doses of at least 0.3 mg (singledose) and 0.5 mg once daily (multiple dose) PF-04457845. Mean fattyacid amide concentrations increased (3.5- to 10-fold) to a plateau andthen were maintained following PF-04457845. FAAH1 activity and fattyacid amide concentrations returned to baseline within 2 weeksfollowing cessation of dosing at doses up to 4 mg. There was noevidence of effects of PF-04457845 on cognitive function. PF-04457845,at doses up to 40 mg single dose and 8 mg once daily for 14 days, waswell tolerated.

CONCLUSIONSPF-04457845 was well tolerated at doses exceeding those required formaximal inhibition of FAAH1 activity and elevation of fatty acid amides.

British Journal of ClinicalPharmacology

DOI:10.1111/j.1365-2125.2011.04137.x

706 / Br J Clin Pharmacol / 73:5 / 706–716 © 2011 Pfizer Limited/Pfizer Inc.British Journal of Clinical Pharmacology © 2011 The British Pharmacological Society

Introduction

A systematic review [1] of placebo-controlled, randomizedcontrolled studies, including five in patients suffering frommultiple sclerosis (MS), comparing either extracts of herbalcannabis or the major psychoactive constituent of can-nabis, D9-tetrahydrocannabinol, demonstrated signifi-cantly greater pain relief from the cannabinoid thanplacebo in the majority of the studies. However, theseagents have also been reported to produce undesirableside effects including impairments in cognition, motorcontrol and predisposition to psychosis, which limit theirutility as therapeutic agents [2].

The integral membrane enzyme fatty acid amidehydrolase-1 (FAAH1) catalyzes the degradation of severalendogenous lipids believed to be neurotransmitters,including N-arachidonyl ethanolamine (anandamide,AEA),N-palmitoylethanolamide (PEA), N-oleoylethanolamide(OEA) and N-linoleoyl ethanolamine (LEA), which havebiological effects via pathways including those initiatedthrough cannabinoid (CB) receptors [3]. Inhibition of thisenzyme has been proposed as a strategy by which theanalgesic properties of cannabinoids may be preservedwithout the accompanying negative effects [4]. Behav-ioural studies comparing FAAH1 (-/-) mice compared withwild-type controls [5–7] indicated attenuated responsesto painful stimuli without disruptions in motility (includ-ing catalepsy and reduced locomotor activity that is com-monly associated with CB1 receptor agonist responses) orbody temperature [5, 7, 8]. Conclusions from geneticallymodified mice are supported by the results frompharmacological manipulation of FAAH1 in rodents (see[9–13]).

PF-04457845 (N-pyridazin-3-yl-4-{3-[5-(trifluoromethyl)-2-pyridyloxy]benzylidene}piperidine-1-carboxamide) is aspecific and irreversible inhibitor of FAAH1 [14]. Oraladministration of PF-04457845 to rats in which pain hadbeen induced using stimuli known to induce hyperalgesicinflammatory (complete Freund’s adjuvant) or non-inflammatory (monosodium iodoacetate) states led toreduced responses that could be interpreted as indicativeof analgesic properties for human chronic pain conditions[15]. Oral administration of PF-04457845 to rats alsoreduced FAAH1 activity and elevated concentrations ofAEA in plasma and brain [15]. Notably near-complete inhi-bition of FAAH1 and maximal elevation of AEA wereinduced by doses of PF-04457845 required for efficacy inthese rodent models.

As a first step in understanding the role of FAAH1 inpain pathways in humans, PF-04457845 has been admin-istered to healthy human subjects and the pharmacokinet-ics (PK), pharmacodynamics (PD) and tolerability profileshave been examined. This manuscript describes threestudies designed to investigate the PK, PD, PK/PD relation-ships and tolerability of single and multiple oral doses ofPF-04457845 in healthy subjects. The relative bioavailabil-

ity of a tablet and solution formulation and the effect offood on the PK of PF-04457845 were also assessed.

Methods

SubjectsEntry criteria for the study population was for healthy maleand/or female subjects aged 18 to 55 years, with a bodymass index of 18 to 30 kg m–2 and a body weight >50 kg.Female subjects had to be of non-childbearing potential(i.e. post menopausal or had undergone a hysterectomyand/or bilateral oophorectomy). All subjects had to bewilling and able to comply with study procedures and allsubjects had to provide written informed consent beforeparticipating. Exclusion criteria included current or historyof clinically significant disease or a history of febrile illnesswithin 5 days of the first dose of study drug, severe acute orchronic medical or psychiatric condition or laboratoryabnormality that may have increased the risk to thesubject or interfered with interpretation of study results, aswell as other standardized criteria concerning allergies, QTc

intervals, treatment with other investigational drugs,alcohol consumption and drug use. Use of concomitantmedications (except ibuprofen), dietary and herbal supple-ments and consumption of grapefruit/pomelo-containingproducts was prohibited throughout the study.The studieswere conducted in accordance with the principles outlinedin the Declaration of Helsinki and Good Clinical Practice. Allstudies were conducted at the Pfizer Clinical Research Unit,Singapore and were approved by the Parkway Indepen-dent Ethics Committee, Singapore.

Study designTwo of the studies were double-blind (Sponsor open),placebo-controlled studies. Sponsor-open means thatsome of the Sponsor’s staff could be unblinded, but at nopoint was any information of any relevance to blindingcommunicated to any investigator or subject. The singlerising dose (SRD) study was a crossover study conducted intwo cohorts of 12 subjects each. In study periods 1 to 3,subjects in cohort 1 received 0.1 mg, 1 mg and 10 mgPF-04457845 with placebo substitution, whilst subjects incohort 2 received 0.3 mg, 3 mg and 20 mg with placebosubstitution. Cohort 2 also had a fourth treatment periodwhere subjects received either 40 mg PF-04457845 orplacebo. Each dose was separated by a minimum of14 days to allow washout of PF-04457845 and to allow forany elevations in FAAH1 to have returned to baseline. Themultiple rising dose (MRD) study was a parallel groupdesign with four cohorts of 10 subjects each (eight active,two placebo). PF-04457845 (0.5 mg, 1 mg, 4 mg and 8 mg)and placebo were administered once daily for 14 consecu-tive days. Doses for the SRD and MRD studies were admin-istered in the fasted state, and PF-04457845 and placebowere administered as oral solutions in both studies. The

Pharmacology and tolerability of PF-04457845

Br J Clin Pharmacol / 73:5 / 707

third study was an open-label, single dose, crossover studyto evaluate the relative bioavailability of PF-04457845tablets compared with solution and to assess the effect offood on the tablet (food effect study). Subjects received, inrandom order, single oral doses of 8 mg PF-04457845 astablets (8 ¥ 1 mg) and as a solution in the fasted state andas tablets following a high fat breakfast. The componentsand timing of the high fat breakfast were in accordancewith the FDA guidance on food-effect bioavailabilitystudies [16].

Dose selectionThe starting dose for the SRD study (0.1 mg PF-04457845)was projected from a preclinical PK/PD model to result in<1% inhibition of the FAAH1 enzyme. The maximum dosein the SRD study was limited by the no observable adverseeffect level in the most sensitive species (dog) from thetoxicology programme, which was associated withAUC(0,24 h) and Cmax values of 6290 ng ml-1 h and482 ng ml-1 respectively. In non-clinical studies (toxicologyand safety pharmacology) with PF-04457845, the malegenital tract, liver and CNS had been identified as the mainpotential target organs (unpublished data). The dose regi-mens selected for the MRD study were aimed at achieving>97% inhibition of FAAH1 activity whilst not exceeding thetoxicokinetic limits.

Pharmacokinetic and pharmacodynamicmethodsBlood samples were collected in the SRD study at intervalsfrom pre dose until 264 h (11 days) after each dose. In theMRD study,pre dose blood samples were collected at inter-vals throughout the study period, whilst post dose sampleswere collected throughout days 1, 7 and 14 and subse-quently at 24–48 h intervals up to the follow-up visit (14–21 days after the last dose). The blood samples in the SRDand MRD studies were split to provide a blood aliquot forleucocyte isolation (for assessment of FAAH1 activity),whilst the remaining sample was used to provide plasmafor analysis of PF-04457845 and the fatty acid amides (AEA,LEA, OEA and PEA). Urine for analysis of PF-04457845 wascollected over 24 h on day 14 of the MRD study. Bloodsamples for analysis of PF-04457845 concentrations onlywere collected in the food effect study at intervals from predose until 96 h post dose.

PF-04457845 was analyzed in plasma and urine usingvalidated LC/MS/MS methods. Details of the analyticalmethods are available in Appendix S1 in the online versionof this article. The calibration range for the plasma andurine assays was 0.100–100 ng ml-1, the lower limit ofquantification was 0.1 ng ml-1, the between-day accuracyranged from -8.8% to 5.3% over the QC range andthe coefficients of variation (CVs) for assay precisionwere �7.7%. PK parameters were calculated using non-compartmental analysis. The following parameters werecalculated as appropriate to each study; maximum plasma

concentration (Cmax), time to Cmax (tmax), minimum plasmaconcentration (Cmin), area under the concentration–timecurve (AUC) to the last measurable time (AUC(0,tlast)), orover the dosing interval (AUC(0,t)) or extrapolated to infin-ity (AUC(0,•)), terminal elimination half-life (t1/2), accumu-lation ratio (Rac; calculated as AUC(0,t)(day 14) : AUC(0,t)(day 1),amount excreted in the urine (Ae) and renal clearance (CLr).

FAAH1 enzyme activity was measured in an ex vivoassay using isolated leucocytes as the source of the FAAH1enzyme [17]. The activity of FAAH1 was measured as theconcentration of deuterated ethanolamine produced fromdeuterated AEA. Samples were assayed for FAAH1 activityusing a validated LCMS/MS method, with a calibrationrange of 10 to 10 000 nM. The between-day assay accuracyfor FAAH1 activity ranged from -18.0% to 24.3% over theQC range and CVs for assay precision were �12.8%.

Plasma samples were analyzed for AEA, LEA, OEA andPEA concentrations using a validated LCMS/MS method[18], with calibration ranges of 0.05 to 10.0 ng ml-1 for AEAand LEA, 0.50 to 100 ng ml-1 for OEA and 1.00 to200 ng ml-1 for PEA. The between-day accuracy for all theassays ranged from -9.0% to 8.6% and CVs for assay pre-cision were �11.5%.

Cognitive function assessmentsCognitive function assessments were conducted in theSRD and MRD studies using an abbreviated CogState/Groton Maze Learning Test (GMLT) battery [19, 20]. Thesetests consisted of computerized tasks that assessed spatialmemory and problem solving (using GMLT), psychomotorfunction (using simple reaction time), attention (usingchoice reaction time), and learning and memory (using theone-card learning test). In addition, GMLT delayed recalltest was included in the MRD study. Subjects had two prac-tice sessions at screening, followed by assessments atintervals throughout the study periods in both studies.

SafetySafety assessments including adverse events monitoring,physical examinations, 12-lead ECGs, vital signs (bloodpressure, pulse rate and body temperature) and laboratorysafety tests were conducted at intervals throughout allstudies. Continuous cardiac monitoring, via telemetry, wasconducted up to 8 h post dose in the SRD study and up to4 h post dose on days 1, 7 and 14 of the MRD study. Allobserved or reported AEs were recorded for all subjectsand AEs were classified as mild, moderate or severeand their relationship to study drug was assessed by theinvestigator.

Statistical analysisSubject numbers for the SRD or MRD studies were chosenbased on the need to minimize exposure of a new chemi-cal entity and the need to have sufficient subjects toenable useful conclusions to be drawn. More formalsample size calculations were not performed for these

G. L. Li et al.

708 / 73:5 / Br J Clin Pharmacol

studies. For the food effect study, a sample size of 12 sub-jects provided 90% confidence intervals (CIs) for the differ-ence between treatments of �0.0905 and �0.2091 on thenatural log scale for AUC(0,•) and Cmax respectively with80% probability. In the MRD study, Rac was analyzed afternatural log transformation using a one way analysis of vari-ance (ANOVA). Means and 90% CIs are presented. For assess-ment of food effect and relative bioavailability, natural log-transformed AUC(0,•), AUC(0,tlast) and Cmax were analyzedusing a mixed effect model. Estimates of the adjustedmean differences and corresponding 90% CIs wereobtained and exponentiated to provide estimates of theratio of adjusted geometric means (test : reference) and90% CIs for the ratios.

The individual outcome measures and a compositechange score for the cognitive function tests were ana-lyzed using mixed linear analysis of covariance (ANCOVA).Analyses and summarization of the CogState endpointswere conducted by CogState (Melbourne, Australia).

Results

SubjectsSeventy-seven healthy subjects were assigned to studytreatment, 24 subjects in the SRD study, 41 in the MRDstudy and 12 in the food effect study. All subjects weremale and aged between 21 and 48 years and the majority(76/77, 99%) were Asian. The demographic and baselinecharacteristics were similar between studies and treat-ment groups. No subjects withdrew from the SRD or foodeffect studies, whilst two subjects withdrew early from theMRD study (neither was due to adverse events).

Pharmacokinetic resultsIn the SRD study, there were supraproportional increases inPF-04457845 exposure following single oral doses rangingfrom 0.1 mg to 10 mg (Table 1, Figures 1 and 2). For singledoses ranging from 10 mg to 40 mg and multiple dosesranging from 1 mg to 8 mg once daily, increases in expo-sure were consistent with increasing dose (Tables 1 and 2,

Figures 1, 2 and 3). Following single and multiple dosing inthe fasted state, PF-04457845 absorption was rapid, withmedian tmax values occurring between 0.5 and 1.2 h postdose. Following Cmax, plasma concentrations exhibited amulti-phasic decline over time with an apparent t1/2 ofapproximately 12 to 23 h. Inter-subject variability was lowto moderate with CVs ranging from 17% to 67% for Cmax

and AUC parameters across both studies. In the MRD study,steady-state appeared to have been achieved by day 7.PF-04457845 exposure increased with once daily dosing,with Rac ratios of approximately 2 to 3. Renal excretionwas negligible with less than 0.1% of the dose beingexcreted in the urine at steady-state. Parameters describ-ing the PK properties of PF-04457845 in healthy humansare presented in Table 1 (SRD study) and Table 2 (MRDstudy).

The relative bioavailability of the tablet formulationcompared with the solution was approximately 100%although the rate of absorption was different. Cmax wasreduced by approximately 35% and delayed by approxi-

Table 1Summary of PF-04457845 pharmacokinetic parameters following single oral dosing (SRD study)

Parameter * (units)PF-04457845 dose0.1 mg 0.3 mg 1 mg 3 mg 10 mg 20 mg 40 mg

N, n 7, 0 9, 4 9, 9 9, 9 9, 9 9, 9 9, 9AUC(0,•) (ng ml-1 h) NC 17.2 (22) 79.4 (30) 372 (21) 1537 (24) 3041 (23) 6048 (17)

AUC(0,tlast) (ng ml-1 h) 0.091 (125) 6.35 (66) 72.6 (32) 366 (21) 1524 (24) 3030 (23) 5969 (18)Cmax (ng ml-1) 0.197 (67) 1.23 (56) 5.96 (37) 28.6 (51) 114 (18) 207 (30) 442 (34)

tmax (h) 0.517 (0.50, 1.00) 1.00 (0.50, 2.00) 1.00 (0.50, 2.00) 1.00 (0.50, 2.00) 1.00 (0.50, 2.00) 1.00 (1.00, 2.00) 1.00 (0.50, 2.00)t1/2 (h) NC 11.6 (23) 15.2 (16) 16.2 (20) 20.5 (30) 21.1 (33) 22.8 (29)

*Geometric mean (%CV) for all except median (min, max) for tmax; arithmetic mean (%CV) for t1/2. N = number of subjects contributing to mean PK parameters; n = number ofsubjects contributing to the mean for AUC(0,•) and t1/2; NC = not calculated.

1000

100

10

1

0.1

0.010 1 2 3 4 5 6 7 8 9 10 11

Time post dose (days)

PF

-044

5784

5 pl

asm

a –1)

conc

entr

atio

n (n

g m

l

Figure 1Mean (SD) PF-04457845 plasma concentrations following single oraldoses (SRD study). 0.1 mg (�), 0.3 mg (�), 1 mg (�), 3 mg (�), 10 mg ( ),20 mg (�), 40 mg (�)

Pharmacology and tolerability of PF-04457845

Br J Clin Pharmacol / 73:5 / 709

mately 1 h when PF-04457845 was administered as tabletscompared with the solution (Table 3). A high fat meal hadno effect on the PK of PF-04457845 when administered astablets with 90% CIs for Cmax and AUC parameters all withinthe range of 80–125% (Table 3).

Pharmacodynamic resultsSingle doses of PF-04457845 appeared to inhibit rapidlythe FAAH1 enzyme with the maximum effect occurringwithin 2 h post dose (Table 4, Figure 4). Almost completeinhibition (>97%) of FAAH1 activity was observed follow-ing 0.3 mg PF-04457845 and above. The FAAH1 activityassay variability was notably lower in samples from sub-jects who had received PF-04457845 than those at base-line or following placebo. It is believed that this is due tothe marked effect of PF-04457845 masking the inherent

variability of the assay, arising primarily from the leucocyteisolation component. The inhibition was maintained forextended time periods following a single dose, the dura-tion of which increased with increasing dose, e.g. FAAH1activity returned to baseline by approximately 168 h(7 days) post dose following 0.3 mg PF-04457845, whilstactivity did not return to baseline until approximately336 h (14 days) post dose following 40 mg PF-04457845.

In the MRD study, all dose regimens (0.5 mg to 8 mgonce daily PF-04457845) demonstrated rapid and sus-tained inhibition of FAAH1 activity (Table 4, Figure 4), withall doses achieving >97% inhibition. Increasing the dose ofPF-04457845 above 0.5 mg once daily did not provideincreased FAAH1 inhibition. This maximal inhibition was

14

12

10

8

6

4

2

00 5 10 15 20 25 30 4035

180160140120100806040200

PF-04457845 dose (mg)

PF

-044

5784

5 m

ean

Cm

ax/d

ose

(ng

ml–1

mg–1

)

PF

-004

5784

5 m

ean

AU

C (

0,µ)/d

ose

(ng

ml–1

h m

g–1)

Figure 2Mean dose normalized Cmax (�) and AUC(0,•) (�) vs.dose of PF-04457845(SRD study)

100

10

1

0.1

0.010 2 4 6 8 10 12 14 16 18 20 22 24

Dosing period

Time post first dose (days)

PF

-044

5784

5 pl

asm

a co

ncen

trat

ion

(ng

ml–1

)

Figure 3Mean (SD) PF-04457845 plasma concentrations following once daily oraldoses (MRD study). 0.5 mg (�), 1 mg (�), 4 mg ( ), 8 mg (�)

Table 2Summary of PF-04457845 pharmacokinetic parameters following single and multiple oral dosing (MRD study)

Parameter * (Units)PF-04457845 dose regimen0.5 mg once daily 1 mg once daily 4 mg once daily 8 mg once daily

Study day 1N 8 8 8 8AUC(0,t) (ng ml-1 h) 21.6 (25) 68.3 (24) 267 (28) 724 (26)Cmax,(ng ml-1) 3.48 (56) 10.6 (28) 34.6 (42) 92.8 (26)tmax (h) 1.00 (0.50, 2.00) 1.00 (0.50, 1.02) 1.00 (0.50, 2.00) 1.00 (0.50, 1.00)

Study day 14N, n 8,7 8,8 7,7 8,8AUC(0,t) (ng ml-1 h) 65.5 (20) 157 (27) 585 (30) 1420 (18)Cmax (ng ml-1) 6.93 (23) 13.9 (29) 53.7 (24) 129 (17)tmax (h) 1.13 (0.50, 1.33) 1.19 (0.53, 2.00) 1.10 (1.08, 2.00) 1.10 (0.50, 1.10)Rac 3.03 (20) 2.30 (22) 2.22 (12) 1.96 (10)t1/2 (h) 18.5 (19) 21.5 (21) 20.0 (25) 21.9 (25)Ae (ng) 0.000 (0) 91.4 (141) 3780 (70) 3206 (57)CLr (ml min-1) ND 0.0008 (5256) 0.0869 (81) 0.0341 (49)

*Geometric mean (%CV) for all except median (min, max) for tmax; arithmetic mean (%CV) for t1/2. Ae = amount excreted in the urine, AUC(0,t) = area under the plasmaconcentration–time profile within the dosing interval of 0 to t (24 h); CLr = renal clearance, Cmax = maximum observed concentration within the dosing interval, CV = coefficient ofvariation, N = number of subjects; n = number of subjects where t1/2 was determined, ND = not determined, Rac = accumulation ratio, t1/2 = terminal half-life, tmax = time to reachCmax.

G. L. Li et al.

710 / 73:5 / Br J Clin Pharmacol

maintained for at least 1 week after the last dose for alldose regimens, with FAAH1 activity not returning to base-line until up to 2 weeks after the last dose up to 4 mg oncedaily.

AEA, LEA, OEA and PEA plasma concentrationsincreased significantly to a plateau following single andmultiple doses of PF-04457845. In the MRD study, theseelevations were similar for each fatty acid amide across alldoses (0.5 to 8 mg once daily PF-04457845; Figure 5) andwere maintained for several days after the last dose on day14. For example, the mean PD endpoints (FAAH1 activityand fatty acid amide concentrations) took 10 to 12 daysfollowing the last dose of 4 mg once daily PF-04457845 toreturn to baseline. On average the concentrations of eachfatty acid amide on day 14 were approximately 10-fold(AEA), 6-fold (OEA), 9-fold (LEA) and 3.5-fold (PEA) higherfollowing PF-04457845 than placebo. Comparison of theAEA time course data with the FAAH1 inhibition datasuggests that at least 97% FAAH1 inhibition was re-quired to maintain elevations of AEA. Examination of therelationship between FAAH1 inhibition and PF-04457845

Table 3Summary of statistical analysis of plasma PF-04457845 exposure (foodeffect study)

Parameter (units) Ratio (%)*

90% confidenceintervalLower Upper

Food effect (comparison: tablet fed vs. tablet fasted [8 mgPF-04457845])

AUC(0,•) (ng ml-1 h) 104 98 111AUC(0,tlast) (ng ml-1 h) 107 97 111Cmax (ng ml-1) 106 90 124

Relative bioavailability (comparison: tablet vs. solution [8 mgPF-04457845])

AUC(0,•) (ng ml-1 h) 103 97 110AUC(0,tlast) (ng ml-1 h) 102 96 110Cmax (ng ml-1) 65 55 76

*Ratio of adjusted geometric means. AUC(0,tlast) = area under the plasmaconcentration–time profile from time zero to the time of the last quantifiableconcentration, AUC(0,•) = area under the plasma concentration–time profile fromtime zero to infinity, Cmax = maximum observed concentration within the dosinginterval.

Table 4Maximum mean % inhibition (from baseline) of FAAH1 activity in human leucocytes (SRD and MRD studies)

SRD study Placebo (n = 21) 0.1 mg (n = 9) 0.3 mg (n = 9) 1 mg (n = 9) 3 mg (n = 9) 10 mg (n=9) 20 mg (n=9) 40 mg (n=9)

FAAH1 % inhibition * 24.3 (38.89) 84.2 (8.06) 96.9 (1.46) 98.6 (0.56) 97.8 (3.44) 99.0 (0.39) 96.1 (5.19) 97.2 (3.54)

MRD studyPlacebo(n = 8)

0.5 mg oncedaily (n = 8)

1 mg oncedaily (n = 8)

4 mg oncedaily (n = 8)

8 mg oncedaily (n = 8)

FAAH1 % inhibition * 52.0 (23.3) 98.2 (0.98) 97.6 (0.45) 98.9 (0.49) 97.2 (1.94)

*Values are means (SD).

0 1 2 3 4 1 2 3 4 5 6 7 8 9 10 11 12 13 14(h) (Days)

Time post dose

120

100

80

60

40

20

0

–20

–40

–60

–80

–100

Mea

n %

inhi

biti

on

of F

AA

H1

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

Time post first dose (days)

150

120

90

60

30

0

–30

–60

–90

–120

–150

–180

Mea

n %

inhi

biti

on

of F

AA

H1

Dosing period

A

B

Figure 4(A) Mean (SD) % inhibition of FAAH1 following single oral doses ofPF-04457845 (SRD study). Placebo (), 0.1 mg (�), 0.3 mg (�), 1 mg (�),3 mg (�), 10 mg ( ), 20 mg (�), 40 mg (�). The SDs have been omittedfrom the placebo line as they were large (average 73%) and obscure theSDs for the active doses. (B) Mean (SD) % inhibition of FAAH1 followingonce daily oral doses of PF-04457845 (MRD study). Placebo (), 0.5 mg(�), 1 mg (�), 4 mg ( ), 8 mg (�)

Pharmacology and tolerability of PF-04457845

Br J Clin Pharmacol / 73:5 / 711

concentration showed that concentrations of at least~1 ng ml-1 PF-04457845 were required to maintain inhibi-tion of FAAH1 activity by at least 97% (Figure 6).

There were no significant differences between placeboand any dose of PF-04457845 for any of the cognitive func-tion tests performed in either the SRD or MRD studieswhether results were examined on an individual test basisor as a composite score (as shown in Figure 7).

SafetyThere were no deaths, other serious or severe adverseevents or discontinuations due to adverse events duringany of the studies. A total of 29 adverse events werereported in the SRD study, of which five were considered tobe treatment-related (one due to placebo and four due toPF-04457845).Somnolence was the only treatment-relatedadverse event to be reported by more than one subject in

the SRD study. A total of 24 adverse events were reportedin the MRD study, of which 18 were considered to betreatment-related (four due to placebo, 16 due toPF-04457845). The only treatment-related adverse eventsin the MRD study reported by more than one subject werehard faeces (six subjects) and headache (three subjects).Only two adverse events were reported in the food effectstudy, of which one adverse event (arthralgia) was consid-ered to be treatment-related. Adverse events reported byat least two subjects in the SRD and MRD studies are pre-sented in Table 5.

No subject or group of subjects showed any clinicallyrelevant change in supine or standing blood pressure (sys-tolic or diastolic) throughout the studies, any evidencefrom vital signs of postural hypotension, or any clinicallyrelevant effects on ECG parameters, including QTc interval(data not shown). There were no treatment-related

Mea

n A

EA

pla

sma

conc

entr

atio

n (

ng m

l–1)

Mea

n O

EA

pla

sma

conc

entr

atio

n (

ng m

l–1)

Mea

n P

EA

pla

sma

conc

entr

atio

n (

ng m

l–1)

Mea

n L

EA

pla

sma

conc

entr

atio

n (

ng m

l–1)

4.5

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

–0.5

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

Dosing period

Dosing period Dosing period

Dosing period

Time post first dose (days)

Time post first dose (days)

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

Time post first dose (days)

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28Time post first dose (days)

A

C D

B

131211109876543210

–1

8

7

6

5

4

3

2

1

0

–1

14

12

10

8

6

4

2

0

Figure 5Mean (SD) elevation of AEA (A), LEA (B), OEA (C) and PEA (D) following once daily oral doses of PF-04457845 (MRD study). Placebo (), 0.5 mg (�), 1 mg (�),4 mg ( ), 8 mg (�)

G. L. Li et al.

712 / 73:5 / Br J Clin Pharmacol

laboratory safety parameters of clinical significance (datanot shown).

Discussion

PF-04457845 is the first inhibitor of FAAH1 that has beendosed to humans. It was well tolerated after single oraldoses from 0.1 to 40 mg, and when administered on

multiple occasions from 0.5 to 8 mg once daily for 14 days.At single doses of 0.3 mg or above or multiple doses of0.5 mg once daily or above, FAAH1 was inhibited nearlycompletely (>97%), with a time course that can bedescribed as fast onset (maximal mean inhibitionobserved within 2 h post dose) and prolonged (FAAH1activity took 1 to 2 weeks to return to baseline dependingon the dose), in accordance with the known tmax of plasmaPF-044587845 exposure and the irreversible mechanism ofinhibition [14, 15].

Despite escalating to both single and multiple doses ofPF-04457845 that achieved maximal effects on inhibitionof leucocyte FAAH1 activity and plasma fatty acid amideconcentrations, PF-04457845 was well-tolerated by all sub-jects to whom it had been dosed. Overall, the nature andincidence of adverse events following PF-04457845appeared similar to placebo and there was no evidence ofany dose-related adverse events. The bland adverse eventprofile is in accordance with the knock-out mouse pheno-type, where the differences from wild type mice were onlyseen in conditions of stress, such as pain stimuli [5–8].

Notably, there was no evidence in these studies ofadverse effects commonly associated with cannabinoids,such as impaired cognition, motor control or psychogeniceffects. The reason for this apparent separation of effectsremains unclear and therefore further investigation wouldbe warranted. Of course, the elevations in fatty acid amides(due to inhibition of FAAH1) may affect mechanisms inaddition to those involved in cannabinoid pathways [21].Evidence of the activation of such mechanisms, such aschanges in appetite or sleep, was not observed in thesestudies from the adverse events profile (with the exceptionof isolated reports of mild somnolence that were not doserelated). However, it should be noted that specific end-points related to such mechanisms were not included inthese studies, although no evidence of sedation could beobserved from the use of a cognitive battery that is moresensitive than voluntary elicitation of adverse events [19].

PF-04457845, an orally available FAAH1 inhibitor, hasdemonstrated variable pharmacokinetic properties inhuman subjects. The distinct dose-dependent Cmax andAUC(0,•) following single dose administration signifies atypical target-mediated drug disposition (TMDD) phe-nomenon, which is commonly observed for large biologi-cal antibodies [22] and less often with small moleculecompounds [23, 24]. The TMDD is most likely to be causedby a capacity-limited irreversible binding of PF-04457845to the FAAH1 enzyme present in the gastrointestinal tractand liver. This explanation is supported by the researchevidence of the distribution of FAAH1 activity and mRNA inthese organs from rat, mouse and humans [25]. This spe-cific binding to FAAH1 is therefore thought to account forthe nonproportional PK of PF-04457845 at lower doses. Athigher doses, this mechanism is saturated and the dispo-sition of PF-04457845 is governed by hepatic metabolismresulting in linear PK.

100

80

60

40

20

00 10 20 30 40 50 60 70

% in

hibi

tio

n o

f FA

AH

1

PF-04457845 plasma concentration (ng ml–1)

Figure 6The relationship between PF-04457845 plasma concentrations andFAAH1 inhibition at doses up to 10 mg PF-04457845 (SRD study), 0.1 mg(�), 0.3 mg (�), 1 mg (�), 3 mg (�), 10 mg ( ). (Higher doses are entirelycompatible with the emergent pattern but are not included as theamount of data at high concentrations obscures the relationship.)

Cha

nge

fro

m b

asel

ine

in c

ogni

tive

func

tio

nst

anda

rdiz

ed c

om

posi

te s

core

Time post dose (h)

4

3

2

1

0

–1

–2

–3

–4

–5Baseline 1 4 8 10 4 8 10 4 8

Day 1 Day 8 Day 14

Figure 7Mean (SE) change from baseline in standardized composite score of thecognitive function tests following once daily oral doses of PF-04457845(MRD study). Placebo (), 0.5 mg (�), 1 mg (�), 4 mg ( ), 8 mg (�)

Pharmacology and tolerability of PF-04457845

Br J Clin Pharmacol / 73:5 / 713

In the MRD study, however, Cmax and AUC(0,t) observedon day 14 appeared to increase proportionally to doseover the range studied. Thus at these doses (0.5 to 8 mg)the nonlinear pathway appeared of little importance indefining the steady-state concentrations. Additional linearpathway(s) appeared to dominate the overall eliminationof the compound at steady-state.

The residual FAAH1 activity in blood leucocytes and theplasma concentrations of the fatty acid amides that wereevaluated in these phase 1 studies have been identified astranslational PD biomarkers i.e. biomarkers that can bemeasured in a similar manner in more than one species toindicate that broadly similar target occupancy and down-stream effects are achieved by PF-04457845 acrossspecies, in this case rat and human. In rodent behaviouralanimal models [15], inhibition of leucocyte FAAH1 activityby approximately 97% was necessary to achieve anti-nociceptive effects, and the same levels of inhibition wererequired to elevate plasma fatty acid amide concentrationsto a maximal degree. The studies reported here confirmthat a similar relationship exists between the degree ofFAAH1 inhibition and the degree of elevation of plasmafatty acid amides in humans.

In comparison with the rapid decline in PK followingsingle or multiple dosing, the PD profiles observed in thesestudies revealed more prolonged effects on FAAH1 inhibi-tion and fatty acid amide elevation than would have

been expected from the PK terminal half-life (~20 h) ofPF-04457845. The visual exploration of the FAAH1 % inhi-bition in relation to PF-04457845 plasma concentrationshowed a typical all-or-nothing steep PK/PD relationship.These observations are in accordance with the irreversibleinhibition mechanism of action [26, 27]. The recovery ofFAAH1 activity is not directly driven by the concentrationof PF-04457845 but probably depends on the slow endog-enous synthesis of the FAAH1 enzyme.

In addition, inhibition of the FAAH1 enzyme resulted insubstantial accumulation of several fatty acid amidesincluding AEA, PEA, OEA and LEA, consistent with the factthat they are known as FAAH1 substrates and FAAH1 isbelieved to be the principal enzyme in their catabolic path-ways [3]. The fact that the maximal extent of fatty acidamide increases in man is independent of dose, similar tothe results observed in rodents, suggests that there is asystem-capped maximal capacity for FAAH1-inhibition byPF-04458745 to produce endocannabinoids. Possibleexplanations for such a profile include the existence ofeither additional catabolic pathways or a negative feed-back mechanism on fatty acid amide synthetic routes[3]. FAAH2 enzyme and N-acyl-ethanolamine (NAE)-hydrolyzing acid amidase (NAAA) have been reported asadditional enzymes that hydrolyze AEA. Unlike FAAH1,FAAH2 is preferentially expressed in select peripheraltissues such as heart and ovary [3]. Unlike FAAH1, which

Table 5Summary of adverse events reported by least two subjects in each study

SRD study

Adverse event

Placebo PF-04457845 doseCohort 1*(n = 9)

Cohort 2†(n = 12)

0.1 mg(n = 9)

0.3 mg(n = 9)

1 mg(n = 9)

3 mg(n = 9)

10 mg(n=9)

20 mg(n=9)

40 mg(n=9)

Number of subjects with adverse events 2 4 1 1 1 2 4 4 1Number of subjects with treatment-related adverse events 0 1 0 0 0 1 2 1 0

Excoriation 1 (0) 0 1 (0) 0 1 (0) 0 2 (0) 0 0Somnolence 0 1 (1) 0 0 0 1 (1) 1 (1) 0 0

Eye pain 0 0 0 0 0 0 0 1 (0) 1 (0)Pyrexia 0 1 (0) 0 0 0 0 0 0 1 (0)

Arthralgia 0 1 (0) 0 0 0 0 1 (1) 0 0Rhinorrhoea 0 1 (0) 0 0 0 0 1 (0) 0 0

MRD study

Adverse eventPlacebo(n = 8)

PF-04457845 dose0.5 mg oncedaily (n = 8)

1 mg oncedaily (n = 8)

4 mg oncedaily (n = 8)

8 mg oncedaily (n = 8)

Number of subjects with adverse events 4 2 3 2 2Number of subjects with treatment-related adverse events 4 2 3 2 1

Faeces hard 2 (2) 2 (2) 2 (2) 0 0Headache 0 2 (2) 0 0 1 (1)

Oropharyngeal pain 1 (0) 1 (0) 0 0 0

Values are numbers of subjects with adverse events (treatment-related adverse events).*Subjects in cohort 1 received placebo, 0.1 mg, 1 mg and 10 mg PF-04457845.†Subjects in cohort 2 received placebo, 0.3 mg, 3 mg, 20 mg and 40 mg PF-04457845.

G. L. Li et al.

714 / 73:5 / Br J Clin Pharmacol

preferentially hydrolyzes AEA as a substrate, NAAA hydro-lyzes PEA much faster than any other NAEs in some condi-tions (see [3]).

Whilst the SRD (first-in-human) study with this irrevers-ible inhibitor was conducted with due caution becauseof the potential for long lasting safety issues, the safetyand tolerability profile observed after administration tohumans was good. FAAH activity returned to baseline after1–2 weeks (depending on dose), indicating that the irre-versible nature of PF-04457845 offered a pharmacologicaladvantage, effectively reducing the dose required for pro-longed elevation of fatty acid amide concentrations.

Overall, the data from these studies support the con-tinued evaluation of PF-04457845 to investigate whetherthe effects of fatty acid amide elevation translate to anal-gesic effects or other effects of benefit to patients. Forsubsequent patient studies, doses of at least 0.5 mg oncedaily should be sufficient, well tolerated and providemaximal effects on FAAH1 activity and fatty acid amideelevation.

Competing Interests

The study was sponsored by Pfizer Worldwide Researchand Development and all authors are current or formeremployees of Pfizer.

The authors would like to thank the staff and subjects whoparticipated in this research, Kuntal Sinha (Pfizer) who wasresponsible for the analysis of PF-04457845, Kathryn Wright(Pfizer) who was responsible for oversight of the analysis ofFAAH1 activity and fatty acid amide concentrations andDeborah Russell (Pfizer) who was responsible for the opera-tional aspects of the studies described here and Paul Maruff(CogState Ltd, Melbourne, Australia) who was responsible forthe cognitive function assessments. The authors would alsolike to thank Samantha Abel (Valley Writing Solutions Ltd) forprofessional medical writing assistance, including prepara-tion of the initial draft based on discussions with the authors,coordination of author review, updates based on author com-ments and management of the journal submission proce-dure. Cogstate Ltd and Valley Writing Solutions Ltd were bothfunded by Pfizer.

REFERENCES

1 Rice ASC, Lever I, Zarnegar R. Cannabinoids and analgesia,with special reference to neuropathic pain. In: SystematicReviews in Pain Research: Methodology Refined, edsMcQuay HJ, Kalso E, Moore RA. Seattle, WA: IASP Press, 2008;233–46.

2 Ware MA, Tawfik BL. Safety issues concerning the medicaluse of cannabis and cannabinoids. Pain Res Manag 2005; 10:(Suppl. A): 31A–7A.

3 Ahn K, McKinney MK, Cravatt BF. Enzymatic pathways thatregulate endocannabinoid signaling in the nervous system.Chem Rev 2008; 108: 1687–707.

4 Ahn K, Johnson DS, Cravatt BF. Fatty acid amide hydrolase asa potential therapeutic target for the treatment of pain andCNS disorders. Expert Opin Drug Discov 2009; 4: 763–84.

5 Cravatt BF, Demarest K, Patricelli MP, Bracey MH, Giang DK,Martin BR, Lichtman AH. Supersensitivity to anandamide andenhanced endogenous cannabinoid signaling in micelacking fatty acid amide hydrolase. Proc Natl Acad Sci USA2001; 98: 9371–6.

6 Lichtman AH, Shelton CC, Advani T, Cravatt BF. Mice lackingfatty acid amide hydrolase exhibit a cannabinoidreceptor-mediated phenotypic hypoalgesia. Pain 2004; 109:319–27.

7 Wise LE, Shelton CC, Cravatt BF, Martin BR, Lichtman AH.Assessment of anandamide’s pharmacological effects inmice deficient of both fatty acid amide hydolase andcannabinoid CB1 receptors. Eur J Pharmacol 2007; 557: 44–8.

8 Huitron-Resendiz S, Sanchez-Alavez M, Wills DN, Cravatt BF,Henriksen SJ. Characterization of the sleep-wake patterns inmice lacking fatty acid amide hydrolase. Sleep 2004; 27:857–65.

9 Boger DL, Sato H, Lerner AE, Hedrick MP, Fecik RA,Miyauchi H, Wilkie GD, Austin BJ, Patricelli MP, Cravatt BF.Exceptionally potent inhibitors of fatty acid amidehydrolase: the enzyme responsible for degradation ofendogenous oleamide and anandamide. Proc Natl Acad SciUSA 2000; 97: 5044–9.

10 Chang L, Luo L, Palmer JA, Sutton S, Wilson SJ, Barbier AJ,Breitenbucher JG, Chaplan SR, Webb M. Inhibition of fattyacid amide hydrolase produces analgesia by multiplemechanisms. Br J Pharmacol 2006; 148: 102–13.

11 Lichtman AH, Leung D, Shelton CC, Saghatelian A,Hardouin C, Boger DL, Cravatt BF. Reversible inhibitors offatty acid amide hydrolase that promote analgesia: evidencefor an unprecedented combination of potency andselectivity. J Pharmacol Exp Ther 2004; 311: 441–8.

12 Kathuria S, Gaetani S, Fegley D, Valino F, Duranti A, Tontini A,Mor M, Tarzia G, La Rana G, Calignano A, Giustino A,Tattoli M, Palmery M, Cuomo V, Piomelli D. Modulation ofanxiety through blockade of anandamide hydrolysis. NatMed 2003; 9: 76–81.

13 Russo R, Loverme J, La Rana G, Compton TR, Parrott J,Duranti A, Tontini A, Mor M, Tarzia G, Calignano A, Piomelli D.The fatty acid amide hydrolase inhibitor URB597(cyclohexylcarbamic acid 3′-carbamoylbiphenyl-3-yl ester)reduces neuropathic pain after oral administration in mice. JPharmacol Exp Ther 2007; 322: 236–42.

14 Johnson DS, Stiff C, Lazerwith SE, Kesten SR, Fay LK, Morris M,Beidler D, Liimatta MB, Smith SE, Dudley DT, Sadagopan N,Bhattachar SN, Kesten SJ, Nomanbhoy TK, Cravatt BF, Ahn K.Discovery of PF-04457845: a highly potent, orallybioavailable and selective urea FAAH inhibitor. ACS MedChem Lett 2011; 2: 91–6.

15 Ahn K, Smith SE, Liimatta MB, Beidler D, Sadagopan N,Dudley DT, Young T, Wren P, Zhang Y, Swaney S,

Pharmacology and tolerability of PF-04457845

Br J Clin Pharmacol / 73:5 / 715

Van Becelaere K, Blankman JL, Nomura DK, Bhattachar SN,Stiff C, Nomanbhoy TK, Weerapana E, Johnson DS, Cravatt BF.Mechanistic and pharmacological characterization ofPF-04457845: a highly potent and selective FAAH inhibitorthat reduces inflammatory and noninflammatory pain. JPharmacol Exp Ther 2011; 338: 114–24.

16 US Department of Health and Human Services, US Food andDrug Administration. Guidance for Industry. Food-effectbioavailability and fed bioequivalence studies. 2002. 1–12.Available from http://www.fda.gov/downloads/RegulatoryInformation/Guidances/UCM126833.pdf (lastaccessed May 2011).

17 Yapa U, Prsakiewicz JJ, Wrightstone AD, Christine LJ,Palandra J, Groeber E, Wittwer AJ. HPLC-MS/MS assay of fattyacid amide hydrolase (FAAH) in blood: FAAH inhibition asclinical biomarker. Anal Biochem 2011. DOI: 10.1016/j.ab.2011.10.042.

18 Palandra J, Prusakiewicz J, Ozer JS, Zhang Y, Heath TG.Endogenous ethanolamide analysis in human plasma usingHPLC tandem MS with electrospray ionization. J ChromatogrB 2009; 877: 2052–60.

19 Collie A, Maruff P, Snyder PJ, Darekar MA, Huggins JP.Cognitive testing in early phase clinical trials: outcomeaccording to adverse event profile in a Phase I study. HumPsychopharmacol 2006; 21: 481–8.

20 Collie A, Darekar A, Weissgerber G, Toh MK, Snyder PJ,Maruff P, Huggins JP. Cognitive testing in early-phase clinicaltrials: development of a rapid computerized test battery andapplication in a simulated Phase I study. Contemp Clin Trials2007; 28: 391–400.

21 Costa B, Comelli F, Bettoni I, Colleoni M, Giagnoni G. Theendogenous fatty acid amide, palmitoylethanolamide, hasanti-allodynic and anti-hyperalgesic effects in a murinemodel of neuropathic pain: involvement of CB1, TRPV1 andPPARg receptors and neurotrophic factors. Pain 2008; 139:541–50.

22 Lobo ED, Hansen RJ, Balthasar JP. Antibodypharmacokinetics and pharmacodynamics. J Pharm Sci2004; 93: 2645–68.

23 Levy G. Pharmacologic target-mediated drug disposition.Clin Pharmacol Ther 1994; 56: 248–52.

24 Gisleskog PO, Hermann D, Hammarlund-Udenaes M,Karlsson MO. The pharmacokinetic modelling of GI198745(dutasteride), a compound with parallel linear and nonlinearelimination. Br J Clin Pharmacol 1999; 47: 53–8.

25 Bisogno T, De Petrocellis L, Di Marzo V. Fatty acid amidehydrolase, an enzyme with many bioactive substrates.Possible therapeutic implications. Curr Pharm Des 2002; 8:125–33.

26 Gisleskog PO, Hermann D, Hammarlund-Udenaes M,Karlsson MO. A model for the turnover ofdihydrotestosterone in the presence of the irreversible5a-reductase inhibitors GI198745 and finasteride. ClinPharmacol Ther 1998; 64: 636–47.

27 Swinney DC. Biochemical mechanisms of drug action: whatdoes it take for success? Nat Rev Drug Discov 2004; 3: 801–8.

Supporting Information

Additional Supporting Information may be found in theonline version of this article:

Appendix S1Analytical method for the analysis of PF-04457845 inK2EDTA plasma

Please note: Wiley-Blackwell are not responsible for thecontent or functionality of any supporting materials sup-plied by the authors. Any queries (other than missingmaterial) should be directed to the corresponding authorfor the article.

G. L. Li et al.

716 / 73:5 / Br J Clin Pharmacol